Abstract : Various physiotherapy and rehabilitation intervention techniques using robots are being actively developed, but there is still much debate about whether to reach the level provided by a human physiotherapist, and at the same time, it is a great motivation for the development of technology. Among the technologies that need to be improved, the natural human-machine interface is an important part of ensuring the safety and participation of patients. In particular, the core technologies are pressure signal measurement through physical contact between the patient and the instrument and recognition of the intention of movement through electromyography measured on the patient’s muscle surface. In this talk, we are going to introduce new approaches both in soft tactile area sensing method and high density Electromyography motion intention extraction method. First, we developed a soft tactile sensing method to cover large areas based on electrical resistance tomography (ERT) and polymer based soft material fabrications. Secondly, we are also going to present a muscle activation estimation method from high density electromyography, which has potential to extract the motion intention with less electromechanical delay and identify the contributions of the muscles which are located deeply from the skin surface.